Sweden's lower birth rate (31.21 against 44.20) was associated with higher ages at marriage and at childbearing: the average age of mothers at the birth of their children was 32.18 years, as against 29.53 years for Honduras in 1966. Late marriage, associated with private property in land, restrained the birth rate in Sweden in a way that is not occurring in Honduras.
With a decidedly lower death rate and higher birth rate Honduras has a very much higher rate of natural increase: projecting the population forward with the respective age-specific birth and death rates we find for Sweden 10 years later 2,478,000, a gain of about 126,000; for Honduras we have 3,398,000, a gain of about 1,000,000. Honduras is having to assimilate somehow a natural increase about eight times as large as that which Sweden had to deal with.
Age differences are also large. In Honduras 51.48 percent of the population in 1966 was under 15 years old; in Sweden in 1800 only 32.60 percent was under 15. The burden of providing schools and other facilities was only a little over 60 percent as great for Sweden as for Honduras, a disadvantage for Honduras even if her economy was able to bear it as well as that of Sweden in 1800.
It is true that Sweden had a greater proportion of people beyond the main working ages: 5.49 percent over 65 in 1800. Honduras has only 1.76 percent over 65 in 1966. But this small disadvantage for Sweden was more than offset by her fewer children. One way of looking at the matter is through the dependency ratio, the number of persons under 15 and over 65 per 100 persons between those two ages. We find that Honduras shows a dependency ratio of 113.89 against Sweden's 61.53. (In this respect Honduras is extreme, but several other less developed countries show dependency ratios higher than 100.)
One point of resemblance between Honduras 1966 and Sweden 1800 was their mortality at older ages. Men of 70 had an expectation of life in Sweden of 6.82 years and in Honduras of 8.63 years. At the youngest ages, on the other hand, the contrast was dramatic: infant mortality (deaths under 1 year of age) was 37.24 per 1,000 live births in Honduras, and 248.96 in Sweden.
Could the difference in completeness of registration be responsible for such a gap? According to the United Nations (1, p. 351) the mortality statistics of Honduras are 75 to 85 percent complete. No evidence is available on the completeness of the early Swedish mortality records, and they could have been better. Even if they were perfect, however, only a small part of this gap would be closed. Let us take extreme assumptions: that 25 percent of the Honduras deaths have been omitted, that the omissions are all infant deaths, and that all omitted deaths have been included among the births, whereas Swedish deaths were 100 percent complete. Even these extreme assumptions give an infant death rate for Honduras less than half that of Sweden.g, however, this is lessh to prove troublesome. control. Dr. Hilton Salhanick has observed that some women practicing the rhythm method will break or lose their thermometers at the critical juncture in theirright, therefore alwayshas some c